Serum IgD and IgE Concentrations in Immunodeficiency Diseases REBECCA H. BUCKLEY and SUSAN A. Fiscus From the Departments of Pediatrics and Microbiology and Immunology, Duke University School of Medicine, Durham, North Carolina 27710
A B S T R A C T Concentrations of IgD and IgE were measured in sera from 165 patients with well-defined immunodeficiency in an effort to find information possibly relevant to the roles of antibodies of these classes in host defense. Values for both immunoglobulins were generally quite low in patients who had marked deficiencies of all three major immunoglobulins, although occasional normal or high normal values for IgD were seen in hypogammaglobulinemic patients. Group mean IgD concentrations were also depressed in patients with Wiskott-Aldrich syndrome and in those with selective IgA deficiency; IgE concentrations were depressed in patients with X-linked immunodeficiency with hyperIgM and in those with ataxia telangiectasia. IgD and IgE were both significantly elevated in patients with extreme hyperimmunoglobulinemia E and undue susceptibility to infection and in a patient with the Nezelof syndrome; none of these patients had histories suggestive of atopy. In addition, the mean IgE concentration was significantly elevated in patients with selective IgA deficiency, many of whom were atopic, and in those with the Wiskott-Aldrich syndrome. The highest IgD concentration (163 mg/100 ml) was found in serum from a boy with variable immunodeficiency who had a lifelong history of severe recurrent pharyngeal infections, primarily streptococcal in etiology. Recurrent staphylococcal infection was a feature common to many but not all patients with elevated IgD and IgE. Depressed cell-mediated immunity was present in many patients with elevated serum IgE concentrations. These data may prove useful in the future delineation of biologic roles for antibodies in these two immunoglobulin classes.
INTRODUCTION Numerous reports have documented a wide variety of abnormalities in concentrations of the three major imReceived for publication 4 June 1974 and in revised form 3 September 1974.
munoglobulins, IgG, IgA, and IgM, in patients with immunodeficiency (1). Less is known, however, about concentrations of IgE in such patients (2-10), and information on IgD concentrations in immunodeficiency can be found primarily only in limited studies (2, 11, 12) or case reports (13). Considerable knowledge has been derived from the study of patients with immunodeficiency diseases about the host defense roles played by various components of the immune system. Since it is unknown whether antibodies of the IgD and IgE classes function in host defense, knowledge of deficiencies or elevations of these proteins in particular groups of infection prone immunodeficient patients may provide some clues as to their biologic function. In the case of IgD antibodies no clearly identifiable biologic activity has yet been found, and indeed, there are only a few reports of antibody activity in molecules having the antigenic properties of IgD (14-17). It is known, however, that IgD molecules do not cross the placenta, are not present in body secretions or urine (12), and do not fix complement or sensitize guinea pig skin for passive cutaneous anaphylaxis (18, 19). IgE antibodies do not cross the placenta or fix complement but are known to be present in external secretions and to bind to basophils and mast cells (20). IgE antibodies so bound effect the release of histamine and the slow reacting substance of anaphylaxis upon combination with antigen, causing both anaphylactic and atopic types of immediate hypersensitivity. In the present study we examined serum IgD and IgE concentrations in 165 patients who represent a spectrum of well-defined immunodeficiency diseases. While these proteins were found generally to be present in low concentrations when there were marked deficiencies in the other three immunoglobulin classes, some immunodeficient patients were found to have elevated concentrations of IgD, IgE, or both. Knowledge of clinical features of several of the latter patients may prove useful in the future delineation of biologic roles for antibodies in these two immunoglobulin classes.
The Journal of Clinical Investigation Volume 55 January 1975-157-165
157
METHODS Study population. Venous blood samples were obtained from 165 patients with well-defined immunodeficiency; the sera were separated by centrifugation and stored at -200C until studied. Most of the patients were seen and/or followed at the Duke University Medical Center from 1967 to 1974, but a few were seen by physicians elsewhere who mailed the patients' sera to us. The deficiency group was composed of 10 cases of infantile X-linked agammaglobulinemia (X Ag),' 15 patients with non-X-linked agammaglobulinemia (Non-X Ag), 8 infants with transient hypogammaglobulinemia (Trans Ag), 3 patients with X-linked immunodeficiency with hyper-IgM (Hyp M), 9 infants with severe combined immunodeficiency (SCID), 79 patients with selective IgA deficiency (A Def), 7 cases of ataxia telangiectasia syndrome (A-T), 3 patients with cellular immunodeficiency with normal or hyperimmunoglobulinemia (Nezelof syndrome or Nez), 4 examples of immunodeficiency with thrombocytopenia and eczema (Wiskott-Aldrich syndrome or W-A) (1), 11 patients with the syndrome of extreme hyperimmunoglobulinemia E and undue susceptibility to infection (Hyp E) (4), 6 patients with other forms of variable immunodeficiency (Var ID), and 10 cases of chronic granulomatous disease of childhood (CGD). Normal controls for serum IgD concentrations consisted of 23 infants, 105 children, and 57 adults; normal controls for serum IgE concentrations included 12 infants, 55 children, and 51 adults. These subjects were selected from hospital personnel and patients attending the Duke Pediatric Outpatient and Cardiac Clinics. Verbal informed consent was obtained for the venous blood collections. Inquiry was made about personal histories of recurrent infection, parasitism, atopic disorders, or autoimmune diseases, and individuals with positive histories were excluded. Measurement of IgD and IgE. Concentrations of IgG, IgA, and IgM were determined by single radial diffusion using antisera specific for each of these human immunoglobulins and primary reference standards prepared in this laboratory (21). IgD concentrations were measured by a similar method using commercial goat anti-human IgD agarose plates (Meloy Laboratories Inc., Springfield, Va.). The secondary IgD reference standard was calibrated against a research IgD standard (67/37) obtained from the WHO Reference Center for Immunoglobulins, Springfield, Virginia. This method was capable of detecting concentrations as low as 1 mg/100 ml. All sera giving rings of precipitate on the goat antihuman IgD and/or IgE (see below) agarose plates were also tested on agarose plates containing either 10, 50, or 100 Al of normal goat serum/15 ml aga-
'Abbreviations used in this paper: A Def, selective IgA deficiency; A-T, ataxia telangiectasia; CGD, chronic gran-
ulomatous disease of childhood; Hyp E, extreme hyperimmunoglobulinemia E and undue susceptibility to infection; Hyp M, X-linked immunodeficiency with hyper-IgM; Nez, Nezelof syndrome or cellular immunodeficiency with normal or hyperimmunoglobulinemia; Non-X Ag, non-X-linked agammaglobulinemia; PS, purified IgE; SCID, severe combined immunodeficiency; T, thymus-derived; Trans Ag, transient hypogammaglobulinemia; Var ID, variable immunodeficiency; W-A, Wiskott-Aldrich syndrome or immunodeficiency with thrombocytopenia and eczema; X Ag, infantile X-linked agammaglobulinemia.
158
R. H. Buckley and S. A. Fiscus
rose to detect antiruminant antibodies which would give false positive rings of precipitate (22) ; all sera reacting on these plates were absorbed with normal goat serum until they no longer gave rings of precipitate and then were retested. IgE concentrations were measured by two methods. Sera were screened initially for high IgE values with commercially available goat antihuman IgE agarose plates (Meloy Laboratories Inc.) which permitted measurement of concentrations as low as 697 U/ml. All sera which had no detectable IgE by single radial diffusion were tested by a modification of the double antibody method of Gleich, Averbeck, and Swedlund (23) which was capable of detecting IgE concentrations less than 1 U/ml. In the early part of this study, the reagents employed in the assay included: (a) sheep anti-IgE (ND) antiserum, kindly provided by Dr. David S. Rowe of the WHO International Reference Center for Immunoglobulins; (b) rabbit antigoat IgG prepared by immunizing 12 rabbits with DEAE cellulose purified goat IgG and repeatedly absorbing the pooled antisera with normal human serum; and (c) purified IgE (PS), isolated by ion exchange and gel filtration chromatography from serum obtained from Dr. 0. Ross McIntyre and labeled with 12"I by the method of Hunter and Greenwood (24). When antiruminant antibodies present in many of the study patients' sera (especially those with either A Def or W-A) were found to give falsely high results in this modification, the method required further revision so as to eliminate ruminant antisera. Assays involving sera containing antiruminant antibodies or sera from patients entering the study during the latter part included the following reagents: (a) rabbit anti-Fc IgE, raised against PS Fc fragment isolated by agarose block electrophoresis of papain-digested, purified PS, (b) pony antirabbit IgG, prepared by immunizing a pony with DEAE cellulose purified rabbit IgG and repeatedly absorbing the antiserum with normal human serum, and (c) purified Bedore IgE myeloma protein, obtained from Dr. Roy Wood of the Immunoglobulin Reference Center, Springfield, Va. and radiolabeled as above. Serum from one of the study patients (B. S.) with extreme hyperimmunoglobulinemia E and undue susceptibility to infection (4) was calibrated against a research standard (68/341) for IgE, obtained from the WHO Reference Center for Immunoglobulins, and found to have 22,300 U/ml. B. S. serum was used as the secondary reference standard in all IgE assays. Because of lack of agreement among various investigators as to the absolute concentration of IgE in the WHO reference standard (25), all IgE data are expressed as units per milliliter, the stated value of 9,346 U/ml in the WHO standard being used as a reference. Statistical nethods. All data were transformed logarithmically for statistical analysis, since concentrations of immunoglobulins G, A, and M have been shown to have a logarithmic gaussian distribution (26). For the logarithmic transformation, undetectable IgD values were called 0.1 mg/ 100 ml and IgE values 1 U/ml. Since both the IgD and IgE concentrations in the control groups were found to have multimodal distributions, however, the differences between group IgD and IgE concentrations were tested by the Mann-Whitney U test for differences of medians (27). Student's t test was also used to test differences between group mean log IgD and IgE concentrations. 95%o confidence intervals were obtained by taking the antilogs of the mean logarithms ±2 pooled SD of the logs of the data, as previously described (21).
TABLE I
Serum IgD Concentrations in Immunodeficiency Patients Geom.
Group
Number
Ages
Range
P value (Mann-
mean
P value (t test)
A B S T R A C T Concentrations of IgD and IgE were measured in sera from 165 patients with well-defined immunodeficiency in an effort to find information possibly relevant to the roles of antibodies of these classes in host defense. Values for both immunoglobulins were generally quite low in patients who had marked deficiencies of all three major immunoglobulins, although occasional normal or high normal values for IgD were seen in hypogammaglobulinemic patients. Group mean IgD concentrations were also depressed in patients with Wiskott-Aldrich syndrome and in those with selective IgA deficiency; IgE concentrations were depressed in patients with X-linked immunodeficiency with hyperIgM and in those with ataxia telangiectasia. IgD and IgE were both significantly elevated in patients with extreme hyperimmunoglobulinemia E and undue susceptibility to infection and in a patient with the Nezelof syndrome; none of these patients had histories suggestive of atopy. In addition, the mean IgE concentration was significantly elevated in patients with selective IgA deficiency, many of whom were atopic, and in those with the Wiskott-Aldrich syndrome. The highest IgD concentration (163 mg/100 ml) was found in serum from a boy with variable immunodeficiency who had a lifelong history of severe recurrent pharyngeal infections, primarily streptococcal in etiology. Recurrent staphylococcal infection was a feature common to many but not all patients with elevated IgD and IgE. Depressed cell-mediated immunity was present in many patients with elevated serum IgE concentrations. These data may prove useful in the future delineation of biologic roles for antibodies in these two immunoglobulin classes.
INTRODUCTION Numerous reports have documented a wide variety of abnormalities in concentrations of the three major imReceived for publication 4 June 1974 and in revised form 3 September 1974.
munoglobulins, IgG, IgA, and IgM, in patients with immunodeficiency (1). Less is known, however, about concentrations of IgE in such patients (2-10), and information on IgD concentrations in immunodeficiency can be found primarily only in limited studies (2, 11, 12) or case reports (13). Considerable knowledge has been derived from the study of patients with immunodeficiency diseases about the host defense roles played by various components of the immune system. Since it is unknown whether antibodies of the IgD and IgE classes function in host defense, knowledge of deficiencies or elevations of these proteins in particular groups of infection prone immunodeficient patients may provide some clues as to their biologic function. In the case of IgD antibodies no clearly identifiable biologic activity has yet been found, and indeed, there are only a few reports of antibody activity in molecules having the antigenic properties of IgD (14-17). It is known, however, that IgD molecules do not cross the placenta, are not present in body secretions or urine (12), and do not fix complement or sensitize guinea pig skin for passive cutaneous anaphylaxis (18, 19). IgE antibodies do not cross the placenta or fix complement but are known to be present in external secretions and to bind to basophils and mast cells (20). IgE antibodies so bound effect the release of histamine and the slow reacting substance of anaphylaxis upon combination with antigen, causing both anaphylactic and atopic types of immediate hypersensitivity. In the present study we examined serum IgD and IgE concentrations in 165 patients who represent a spectrum of well-defined immunodeficiency diseases. While these proteins were found generally to be present in low concentrations when there were marked deficiencies in the other three immunoglobulin classes, some immunodeficient patients were found to have elevated concentrations of IgD, IgE, or both. Knowledge of clinical features of several of the latter patients may prove useful in the future delineation of biologic roles for antibodies in these two immunoglobulin classes.
The Journal of Clinical Investigation Volume 55 January 1975-157-165
157
METHODS Study population. Venous blood samples were obtained from 165 patients with well-defined immunodeficiency; the sera were separated by centrifugation and stored at -200C until studied. Most of the patients were seen and/or followed at the Duke University Medical Center from 1967 to 1974, but a few were seen by physicians elsewhere who mailed the patients' sera to us. The deficiency group was composed of 10 cases of infantile X-linked agammaglobulinemia (X Ag),' 15 patients with non-X-linked agammaglobulinemia (Non-X Ag), 8 infants with transient hypogammaglobulinemia (Trans Ag), 3 patients with X-linked immunodeficiency with hyper-IgM (Hyp M), 9 infants with severe combined immunodeficiency (SCID), 79 patients with selective IgA deficiency (A Def), 7 cases of ataxia telangiectasia syndrome (A-T), 3 patients with cellular immunodeficiency with normal or hyperimmunoglobulinemia (Nezelof syndrome or Nez), 4 examples of immunodeficiency with thrombocytopenia and eczema (Wiskott-Aldrich syndrome or W-A) (1), 11 patients with the syndrome of extreme hyperimmunoglobulinemia E and undue susceptibility to infection (Hyp E) (4), 6 patients with other forms of variable immunodeficiency (Var ID), and 10 cases of chronic granulomatous disease of childhood (CGD). Normal controls for serum IgD concentrations consisted of 23 infants, 105 children, and 57 adults; normal controls for serum IgE concentrations included 12 infants, 55 children, and 51 adults. These subjects were selected from hospital personnel and patients attending the Duke Pediatric Outpatient and Cardiac Clinics. Verbal informed consent was obtained for the venous blood collections. Inquiry was made about personal histories of recurrent infection, parasitism, atopic disorders, or autoimmune diseases, and individuals with positive histories were excluded. Measurement of IgD and IgE. Concentrations of IgG, IgA, and IgM were determined by single radial diffusion using antisera specific for each of these human immunoglobulins and primary reference standards prepared in this laboratory (21). IgD concentrations were measured by a similar method using commercial goat anti-human IgD agarose plates (Meloy Laboratories Inc., Springfield, Va.). The secondary IgD reference standard was calibrated against a research IgD standard (67/37) obtained from the WHO Reference Center for Immunoglobulins, Springfield, Virginia. This method was capable of detecting concentrations as low as 1 mg/100 ml. All sera giving rings of precipitate on the goat antihuman IgD and/or IgE (see below) agarose plates were also tested on agarose plates containing either 10, 50, or 100 Al of normal goat serum/15 ml aga-
'Abbreviations used in this paper: A Def, selective IgA deficiency; A-T, ataxia telangiectasia; CGD, chronic gran-
ulomatous disease of childhood; Hyp E, extreme hyperimmunoglobulinemia E and undue susceptibility to infection; Hyp M, X-linked immunodeficiency with hyper-IgM; Nez, Nezelof syndrome or cellular immunodeficiency with normal or hyperimmunoglobulinemia; Non-X Ag, non-X-linked agammaglobulinemia; PS, purified IgE; SCID, severe combined immunodeficiency; T, thymus-derived; Trans Ag, transient hypogammaglobulinemia; Var ID, variable immunodeficiency; W-A, Wiskott-Aldrich syndrome or immunodeficiency with thrombocytopenia and eczema; X Ag, infantile X-linked agammaglobulinemia.
158
R. H. Buckley and S. A. Fiscus
rose to detect antiruminant antibodies which would give false positive rings of precipitate (22) ; all sera reacting on these plates were absorbed with normal goat serum until they no longer gave rings of precipitate and then were retested. IgE concentrations were measured by two methods. Sera were screened initially for high IgE values with commercially available goat antihuman IgE agarose plates (Meloy Laboratories Inc.) which permitted measurement of concentrations as low as 697 U/ml. All sera which had no detectable IgE by single radial diffusion were tested by a modification of the double antibody method of Gleich, Averbeck, and Swedlund (23) which was capable of detecting IgE concentrations less than 1 U/ml. In the early part of this study, the reagents employed in the assay included: (a) sheep anti-IgE (ND) antiserum, kindly provided by Dr. David S. Rowe of the WHO International Reference Center for Immunoglobulins; (b) rabbit antigoat IgG prepared by immunizing 12 rabbits with DEAE cellulose purified goat IgG and repeatedly absorbing the pooled antisera with normal human serum; and (c) purified IgE (PS), isolated by ion exchange and gel filtration chromatography from serum obtained from Dr. 0. Ross McIntyre and labeled with 12"I by the method of Hunter and Greenwood (24). When antiruminant antibodies present in many of the study patients' sera (especially those with either A Def or W-A) were found to give falsely high results in this modification, the method required further revision so as to eliminate ruminant antisera. Assays involving sera containing antiruminant antibodies or sera from patients entering the study during the latter part included the following reagents: (a) rabbit anti-Fc IgE, raised against PS Fc fragment isolated by agarose block electrophoresis of papain-digested, purified PS, (b) pony antirabbit IgG, prepared by immunizing a pony with DEAE cellulose purified rabbit IgG and repeatedly absorbing the antiserum with normal human serum, and (c) purified Bedore IgE myeloma protein, obtained from Dr. Roy Wood of the Immunoglobulin Reference Center, Springfield, Va. and radiolabeled as above. Serum from one of the study patients (B. S.) with extreme hyperimmunoglobulinemia E and undue susceptibility to infection (4) was calibrated against a research standard (68/341) for IgE, obtained from the WHO Reference Center for Immunoglobulins, and found to have 22,300 U/ml. B. S. serum was used as the secondary reference standard in all IgE assays. Because of lack of agreement among various investigators as to the absolute concentration of IgE in the WHO reference standard (25), all IgE data are expressed as units per milliliter, the stated value of 9,346 U/ml in the WHO standard being used as a reference. Statistical nethods. All data were transformed logarithmically for statistical analysis, since concentrations of immunoglobulins G, A, and M have been shown to have a logarithmic gaussian distribution (26). For the logarithmic transformation, undetectable IgD values were called 0.1 mg/ 100 ml and IgE values 1 U/ml. Since both the IgD and IgE concentrations in the control groups were found to have multimodal distributions, however, the differences between group IgD and IgE concentrations were tested by the Mann-Whitney U test for differences of medians (27). Student's t test was also used to test differences between group mean log IgD and IgE concentrations. 95%o confidence intervals were obtained by taking the antilogs of the mean logarithms ±2 pooled SD of the logs of the data, as previously described (21).
TABLE I
Serum IgD Concentrations in Immunodeficiency Patients Geom.
Group
Number
Ages
Range
P value (Mann-
mean
P value (t test)
